Couples in which both members smoke represent key targets for smoking cessation interventions. Quit rates are lower and relapse rates are higher among these individuals, and each person's health risks are amplified due to exposure to their partner's smoking. As a strategy to promote quit, and despite these facts, message framing interventions have not highlighted how smoking outcomes related to health, appearance and finances might affect both partners in a couple jointly. For couples in which both partners smoke, intervention messages can highlight how positive outcomes of quitting (i.e., positive frame) or costs of not quitting (i.e., negative frame) affect the individual smoker or the couple. This study extends prior framing research by examining how gain- versus loss-framed messages focused on individual- versus couple-level outcomes influence (a) message acceptance, operationalized by evaluations of message accuracy, credibility and outcome relevance; (b) likely mediators of successful cessation at both individual and relationship levels, including intentions, risk perceptions, self- and dyadic efficacy, and communal coping operationalized both by self-report and first person plural pronoun use (we-talk) in a couple conversation about quitting; and (c) smoking behavior change over 6 months, verified biochemically. Participants will be 120 couples (married or living as married, same and opposite sex) in which both members smoke and neither plans to quit during the next 30 days. We will randomize couples to 1 of 4 cells in a 2 x 2 factorial design that crosses gain vs. loss frame with self vs. couple outcomes focus. The specific intervention manipulations will be implemented and reinforced via written narratives, summaries of key framing points, and booster messages given 1 week after the initial intervention session. In addition to a baseline assessment before randomization, couples will complete an immediate post-intervention survey assessing message acceptance plus hypothesized individual and relational mediators; participate in a joint post-intervention interview about smoking (with transcripts later yielding we-talk via automatic text analysis); and provide follow-up data 1 and 6 months later to assess outcomes such as intentions to quit, amount smoked, and cessation. We hypothesize that (a) framed messages focused on couple- rather than individual-level outcomes will be more effective in motivating quitting; (b) gain-framed messages will be more persuasive than loss-framed messages, especially when the message outcome focus is the couple; (c) couple-focused messages will be most effective in prompting hypothesized relational change mechanisms such as dyadic efficacy and communal coping; and (d) all these associations will be stronger in couples with high relationship quality. If results provide preliminary support for these novel and uninvestigated hypotheses, effect size estimates will be used to propose a larger randomized trial. This research should add important strategic insights to new applications of message framing in clinical practice and public health when the high-risk target group is couples in which both partners smoke.